Wireless Performance

By TG Publishing Team, published on January 3, 2003
Source: Tom's Guide US | Keywords: , , ,

5. Wireless Performance

Since the table above shows the wireless performance results, let's start there. The table data shows that the 2521's performance is pretty consistent across all four test conditions. This was suprising, given the little printed circuit antenna in the AP's Compact Flash radio!

But I suspect part of the reason is the strong 802.11b radio in the Atheros-based NETGEAR WAB501 Dual-Band card that I've started to use as my standard test client. But numbers is numbers, and it was also nice to see that there was no throughput loss when I enabled 128bit WEP.

Figure 6: Wireless Throughput
(click on the image for a full-sized view)

Figure 6 shows 30 second NetIQ Chariot runs for all four test conditions, which reveal about a 30% cyclical throughput variation.

Powerline Performance

Turning to the powerline side, I again used Chariot and Qcheck to run my Powerline four-condition test group, using two 2502 Powerline Ethernet Adapters to construct the test network. The numbers please...

Powerline Performance Test Results

Test Conditions:
See this page.

Firmware/Driver Version:
No info

Test Description

Transfer Rate (Mbps)

[1 MByte data size]

Response Time (msec)

[10 iterations 100Byte data size]

UDP stream
[10S@500 kbps]

(Actual throughput- kbps)

(Lost data- %)

Condition 1

5.2

3(avg)
4(max)

467

0%

Condition 2

3.8

2(avg)
3(max)

475

0%

Condition 3

3.4

2(avg)
3(max)

424

0%

Condition 4

3.2

3(avg)
3(max)

420

0%

[Details of how we test can be found here.]

These results show that throughput fell off pretty quickly once my test powerline network got outside the same room. To put things in perspective, a 10BaseT Ethernet connection will give you around 8Mbps, and it won't vary with distance (assuming you use good CAT5 cable).

So, in the best case, my best case throughput was about 65% of what a 10BaseT Ethernet connection would yield, with performance over a typical residential powerline network distance dropping to as low as 40% of Ethernet performance. I also did Chariot runs to look at throughput variation, with the results shown in Figure 7.

Figure 7: Powerline Throughput - Two SS2502
(click on the image for a full-sized view)

A look back

Since this was my first test of a product that uses Intellon's second-generation chipset, I checked back to see how my results compared with a product based on their first-generation 5130 chipset. Figure 8 shows the Chariot runs from my Linksys PLEBR10 [reviewed here] testing.

Figure 8: Powerline Throughput - Two Linksys PLEBR10
(click on the image for a full-sized view)

As you can see, the results are quite different, with three out of four test locations achieving better than 5Mbps average thoughput under the same test conditions. I hope that these results aren't indicating any fundamental limitations in the newer chipset, and that performance can be improved with some driver tweaking.

I also plugged a hairdryer into the same outlet as one of the powerline adapters and did some Chariot runs while cycling the dryer on and off. I didn't see any throughput changes that I could correlate to the hairdryer's operation.

Comments | Print | Send to a friend

Sponsored links

Comments

Anonymous 12/27/2007 3:33 AM
Hide
-0+

Does it have the Vista driver ?
SS2521

Comments are closed on this page.

Sponsored links